MPM Space Sciences - Machine Learning

• MPM Space Sciences - Machine Learning
  • Preliminaries
    • Python
      • Linux - Ubuntu & Mint
      • Mac OS
      • Windows
    • Virtualenv
    • Virtualenv-wrapper (Optional)
      • Ubuntu, Mint, and Mac OS
      • Windows
    • Editor
  • Setup

Preliminaries

Install and test Python, Virtualenv and Virtualenv-wrappers on your machine. We suggest to work with Python 3.8 or higher versions.

Python

Here, we report minimal instructions. See this guide for details.

Linux - Ubuntu & Mint

Open your terminal and first check which Python version is installed on your machine:

python --version

python3 --version

If the version does not match the requirements, run the following command:

sudo apt-get update
sudo apt-get install python3.8 python3-pip

or

sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt-get update
sudo apt-get install python3.8 python3-pip

After the installation, you should be able to open a Python shell:

 ~/veos_digital/repos/mpm_space_sciences/python3                                                                                                   
Python 3.9.1 (default, Dec 10 2020, 11:11:14) 
[Clang 12.0.0 (clang-1200.0.32.27)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> 

Exit by pressing Ctrl + D

Mac OS

Open your terminal and first check which Python version is installed on your machine:

python --version

python3 --version

If the version does not match the requirements, first install Homebrew by following these instructions. Then, update homebrew by running:

brew update && brew upgrade

Finally, install Python:

brew install python3

After the installation, you should be able to open a Python shell:

 ~/veos_digital/repos/mpm_space_sciences/python3                                                                                                   
Python 3.9.1 (default, Dec 10 2020, 11:11:14) 
[Clang 12.0.0 (clang-1200.0.32.27)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> 

Windows

Open PwerShell, then check which Python version is installed on your machine:

python --version

If the version does not match the requirements, open the Microsoft Store app, search for Python and choose the highest version available.

Virtualenv

You can install Virtualenv through pip:

sudo pip3 install virtualenv 

You can now create a virtualenv running:

virtualenv venv_name 

or specifying a Python interpreter, e.g.,

virtualenv -p /usr/bin/python3 venv_name

You can now activate the virtual environment by running:

source venv_name/bin/activate

run deactivate to exit the virtual environment.

Virtualenv-wrapper (Optional)

Ubuntu, Mint, and Mac OS

Run

sudo pip install virtualenvwrapper

Then, on Mac OS, Ubuntu, and Mint add to your shell startup file (.bashrc, .profile) the following lines:

export WORKON_HOME=$HOME/.virtualenvs
export PROJECT_HOME=$HOME/Devel
source /usr/local/bin/virtualenvwrapper.sh

Save, exit, and reload the startup file, e.g.,

source ~/.bashrc

Now try creating a virtual environment by running:

mkvirtualenv --python=python3 mpm_space_sciences                                   

Windows

Install virtualenvwrapper-win and follow the instruction on the website.

Thereafter, try creating a virtual environment by running:

mkvirtualenv --python=python3 mpm_space_sciences                                   

Editor

You can use your preferred editor. If you do not have one we suggest Visual Studio.

Setup

After successfully installing Python and Virtualenv,

1. let us create a virtual environment called mpm_space_sciences. Follow the instructions above, if you do not know how to do it.

2. Clone this repository:

   git clone ...

3. Install it:

   cd mpm_space_sciences
   pip install -e ./

   Where ./ means that pip will use the ./setup.py script to install the packagge, and the -e option allows you to modify the source code of the package you installed.